Forgive me if I mess up my chemistry/physics here:
From what I understand H2O has a tremendous capacity to absorb heat. We dont use pure water in a car radiator because it’s corrosive and it freezes. Also I think there’s something about raising the boiling point although that may just be due to the system being pressurized.
Is there some other man made fluid that has a greater heat capacity than water? Maybe just some sort of chemistry curiosity that would be too expensive for practical use?
If not, is it understood why water is the best? Is it understood on a physics level why this particular molecule works this way? I know water has other interesting characteristics as well, expands as it freezes,etc. Why?
I realize this is probably one of those ‘why’ questions whose ultimate answer is “you need to take a physics class” but a laymans explanation will do for now or at least help me think about the question properly.
Why frozen water expands: The hydrogen forms “caverns” which are partly empty and therefore increase volume for a given weight.
As for coolant, it depends what you want to do. Carbon dioxide is neutral enough as a fluid and can easily be made supercritical which helps transfer energy. Molten salts may be solid at room temperature but they can be fluid when heated up.
As for the use of water, there is definitely a physics aspect but also an economics one: Light water’s cheap, easily available and can be cheaply purified.
In nuclear reactors, the first reactors were made with water as coolant I believe. Since nuclear reactor design is both cost-conscious and risk averse, there is a tendency to stick with what is cheap and previously used.
Yes. Sodium is a better coolant than water in nuclear reactors, because it boils at a higher temperature. Well. “Better” in terms of performance - it has the major drawback of exploding when it comes in contact with water, which is superbad because actual reactor designs use water in the outer coolant loop and sodium to cool the core, and a heat exchanger where there is water on one side and sodium on the other. If the heat exchanger leaks, the obvious will happen. Also it’s super corrosive.
So from the perspective of making a reliable nuclear reactor, it’s not as good a coolant.
We don’t use water because it freezes at too high of a temperature and boils at too low of a temperature. A mix of water and ethylene glycol has a much lower freezing point and a much higher boiling point. So while water has good thermal conductivity and can hold a great deal of heat (this is called its thermal capacity), its operating temperature range isn’t all that fantastic.
There are other variations on this theme like propylene glycol, which is again a water and glycol type of mix used to extend the operating range of the coolant. The thermal capacity is reduced a bit, but it’s usually still good enough for the application.
Oil is also used as a coolant. A lot of folks say that VW Bugs are air cooled, and in a way they are, but really they are oil cooled. It’s the oil that transports the heat out of the engine block to where the fan is located. The same oil also provides lubrication for the engine.
Large electrical transformers also often use oil cooling.
What is the “best” coolant often depends on what is being cooled. In a lot of ways, water is hard to beat. It’s cheap, readily available, isn’t toxic to humans (in reasonable quantities), and generally doesn’t create an environmental disaster if it spills. On the other hand, its operating range sucks compared to glycols and alcohols, it is electrically conductive (which is very bad when trying to cool certain types of things), and it reacts with a lot of metals
As for heat capacity alone, ammonia has a higher heat capacity than water. I think it should be fairly obvious why we don’t use ammonia as a general coolant in the same way that we use water.
The neat thing is we can engineer around water’s limitations in a whole lot of ways, too, by controlling things such as the flow rate, pressure, type of flow, and so on. We can filter the cooling system or even deionize the water to limit electric conductivity. We can chill the water to increase its cooling capacity. Additives can increase its operating range. In the end, water is virtually free, mostly safe, and nearly anyone can work on its supporting infrastructure safely.
Can’t control the lack of specific heat or the low boiling point past a certain level, of course. Just saying that when you absolute need the performance, water won’t cut it. Like for making a nuclear submarine truly haul ass, or a nuclear aircraft (the Russians were looking into it. they changed their minds for obvious reasons but they had a paper design), or a nuclear-electric spacecraft or nuclear powered satellite.
It’s cold, obviously. But it also demonstrates a phenomenon called second sound, where heat is transferred in waves instead of by particles randomly colliding with their neighbors. This gives liquid helium a very high thermal conductivity. A hotspot in a tank of liquid helium will get averaged out in fractions of a second.
Liquid helium also has a very high heat capacity per unit mass, though its low density means that the volumetric density is not so hot.
Actually, liquid helium is terrible as a coolant. Its heat capacity per unit mass may be pretty good, but its mass is so small that you need a lot of it to cool something down. Its heat capacity per unit volume is terrible. You have to boil off a LOT of it to cool down anything else.
I speak from experience here – I’ve used liquid helium to cool down specimens. When you start cooling down you use liquid nitrogen to get your temperature lower, not only because LN is good at higher temperatures, biut also because it’s a LOT cheaper, and because its heat capacity per unit volume is much better than helium’s. The only reason people use liquid helium at low temperatures is that it’s about the only thing that will get you that cold. (Hydrogen will get you almost as low, but when it boils off yuou fill the room with highly flammable hydrogen. And hydrogen’s heat capacity per unit volume is as pitiffuil as helkium’s).
Much better, if you don’t have to go all the way down to helium’s boiling point, to use a helium refrigerator*. You re-use the helium, so it doesn’t all go boiling off. You don’t need as much, and it’s a lot more convenient. It’ll get you to 14 K quickly and easily.
*Not to be confused with a Helium Dilution Refrigerator, which is a whole different beast, and much more complex. but it’ll get you below 4.2K.
It wasn’t just tinkering on paper. They had an actual test aircraft fitted with a nuclear reactor. They made a lot of flights without the reactor being powered just to test the plane’s flight characteristics when loaded with the full weight of a reactor and all of the shielding necessary so that it didn’t kill the crew. They also did tests with the reactor running, experimenting with shielding thicknesses and radiation levels in various crew compartments and such. They stopped long before they reached the stage of actually using the reactor to drive the plane’s engines.
And they weren’t alone. The U.S. basically did the same thing, and also stopped long before actually driving the engines. And actually, the U.S. version came first. The U.S. version flew from 1955 to 1957, and the Soviet version flew from1961 to 1969.
I didn’t say it was cheap :). But if you have a lot of it, it should have virtually infinite capacity to cool your specimen because of the second sound effect. That is, as long as you keep it in its superfluid phase, which means that you have to stay well below the boiling point (2.2 K instead of 4.2 K).
Of course you can increase you ability to transfer heat with any liquid by increasing the flow rate–forced convection instead of conduction. But that’s not so hot if your sample is delicate, say.
Heat capacity is not the only criterion. It’s certain helpful, but you can compensate for a lower heat capacity with a higher flow rate of coolant.
Two other properties of water, besides its economy and safety, are also relevant: its very high thermal conductivity and latent heat. Both are near the very top for liquids. Thermal conductivity means heat can be transferred to and from and within the water quickly, and with smaller temperature gradients. A high latent heat means it takes a lot of heat to boil the stuff. If you want your system kept liquid, that can be helpful, because it means the water has to absorb a tremendous amount of heat before it starts to boil. On the other hand, sometimes you use the boiling to do the cooling, and there the large latent heat means you get a lot of cooling for boiling off a fairly modest amount of water.
As for why water has all these properties, the answer is its hydrogen-bonding network, which make liquid water much less different from a networked solid (like diamond or silica) than the fact that it’s a liquid would make you think. So it has some of the properties of a networked solid, e.g. a relatively high (for its molecular weight) boiling point, high latent heat and heat capacity, high thermal conductivity.
What actually needs explaining is how water can do all of these things but still be as liquid and mobile as it is. By rights it should be an exceedingly viscous fluid, if not a glass. The answer to that is much more subtle: it seems to rest on the extreme lightness of the hydrogen atom, which allows some very complex and rapid alterations in the hydrogen bonded network, quite possibly the result of quantum tunneling events. It’s a subject of theoretical study.
Actually, a fair bit of water’s favorable properties are just due to the fact that it has small molecules, and yet remains non-gaseous under reasonable circumstances. Specific heat is often expressed on a per-mass or per-volume bases, and those numbers are all over the place, but it can also be expressed on a per-molecule basis, and that’s much more consistent from one substance to another. So a substance with small molecules like water can fit a lot more molecules into any given mass or volume, and so has more heat capacity for that mass or volume.
“Dihydrogen Monoxide the Magnificent.”
Respective of nothing much, even with a chemistry background I have trouble thinking of ammonia as a substance by itself. It’s mixture of [del]water[/del] DHM and a little other stuff… right? Right? :rolleyes:
The highest that I know of. If you just want a coolant, high heat capacity is good for you. Others have pointed out the problems with water and why we’ll often use other substances that have lower heat capacities.
If you want a refrigerant, then you’ll need to consider the thermodynamics of the fluid’s phase changes.
I wonder if that would improve safety in land-based reactors. Lead has a lower vapor pressure, so you can’t get steam explosions like what blew up the Chernobyl plant. It doesn’t devolve to hydrogen and oxygen at high temperatures and/or with a catalyst. When you have a meltdown, the resuling “corium”, the melted remnants of the reactor, fuel, and other metal might be safer because the hot fuel is mixed with lead which blocks a lot of the gamma.
Any safety procedures at all, even just the (one would think) obvious one of “don’t actively try to trash the reactor”, are sufficient to prevent what happened at Chernobyl.
